Review Summarization Using Keyword Polarity and Self-Critique
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users experience cognitive overload and decision-making fatigue when navigating through numerous reviews on network-based interfaces, leading to potential abandonment of interactions with items and interfaces.
Innovation Solution
A system utilizing a processor and non-transitory memory to select and rank target keywords based on polarity, generate summaries, and iteratively refine summaries using a self-critique module to enhance quality, ultimately displaying a concise summary on a user interface.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If hundreds of individual reviews are displayed to provide comprehensive information, then information completeness is improved, but user cognitive overload increases
Solution Approach 1:
The patent segments the large set of hundreds of reviews into smaller, manageable groups based on common themes, features, or sentiment categories. Each segment is represented by a summary that captures key points, allowing users to access comprehensive review information in organized chunks rather than overwhelming them with all individual reviews at once.
Solution Approach 2:
The patent extracts essential information from numerous individual reviews to create condensed summary representations. These summaries capture the core sentiments, key features mentioned, and overall ratings without requiring users to read through every single review, thereby preserving important information while reducing cognitive load.
2Reliability
If all review data is presented to ensure thorough evaluation, then decision accuracy is improved, but time consumption increases
Solution Approach 1:
The patent performs preliminary processing of review data by automatically generating summaries, identifying key themes, and organizing reviews before they are presented to users. This pre-processing work is done in advance, so when users view the summaries, they get distilled essential information quickly without having to manually read and synthesize hundreds of individual reviews themselves.
Solution Approach 2:
The patent creates simplified copy or representation of the essential content from numerous reviews in summary form. These summaries serve as condensed versions that capture the meaningful information from all reviews, allowing users to make accurate decisions by reading the summaries rather than the full original review set.
3Measurement precision
If detailed review analysis is performed to improve summary quality, then summary accuracy is improved, but processing complexity increases
Solution Approach 1:
The patent implements self-service mechanisms where the system automatically performs review analysis, theme identification, and summary generation without requiring manual intervention. The processing system uses automated algorithms to extract key information and generate accurate summaries, reducing the need for complex manual processing while maintaining high summary quality through systematic automated analysis.
Data Source
AI summary
Systems and methods for generating aggregate reviews are disclosed. Generating aggregate reviews includes receiving review data associated with an item object in a database and selecting a subset of the review data based on a first selection criteria. The systems and methods further include generating, a target keyword score for each respective target keyword based on a polarity of the target keywords, selecting a subset of the target keywords based on a predetermined range of the target keyword scores, generating one or more target keyword summaries for each respective target keyword, generating an item object summary based on the one or more target keyword summaries, iteratively modifying the item object summary using a self-critique module, and generating a set of instructions to cause the item object summary to be displayed on a user interface.